Summary The aim of this proposal is to develop the method of femtosecond (fs) crystallography for the structure determination of membrane proteins, where X-ray structure analysis is based on hundreds of thousands of X- ray diffraction patterns from a steam of fully hydrated nano/ microcrystals of membrane proteins, collected using the new high energy fs X-ray laser at LCLS in Stanford. The LCLS started its operation in the fall of 2009 and provides fs-pulses of an intensity that exceeds third-generation synchrotron sources by 12 orders of magnitude. Membrane proteins are of extreme importance in all living cells as they catalyze vital functions like respiration, photosynthesis, transport, and cell communication. 30% of all human proteins are membrane proteins and more than 60% of all drugs are targeted to membrane proteins. Despite their extreme importance, the understanding of their molecular function is hampered by the lack of structure information; while more than 60,000 structures of soluble proteins have been solved by X-ray crystallography and NMR, less than 250 different membrane protein structures have so far been determined. The determination of membrane protein structures solved to date often involved a time-consuming process where it took years (or sometimes even decades) to grow large, well-ordered crystals suitable for X-ray structure determination. Furthermore, X-ray-induced radiation damage is a major problem for many membrane protein crystals, especially when they contain metals and/or redox active cofactors. The X-ray-induced radiation damage imposes a limitation for X-ray diffraction on microcrystals, even under cryogenic conditions. This proposal is based on the first proof of principle for fs-nanocrystallography by the collection of 3 million diffraction patterns on nano/microcrystals of the membrane protein Photosystem I in December 2009 at LCLS, using fs X-ray pulses. Photosystem I, which served as the model system, has a molecular weight of 1,056,000 Daltons and consists of 36 proteins and 381 cofactors that are non-covalently bound, making Photosystem I one of the most complex membrane proteins that has been crystallized to date. These experiments have already proven that the diffraction before destroy principle, first shown in 2006 for an image etched into a silicon-nitrate film, (Chapman 2006, Nature Physics), can be directly extended to one of the most fragile protein crystals that exists to date, which contain 78% solvent and only 4 salt bridges involved in crystal contact. This proposal aims to open an exciting new avenue for membrane protein crystallography, where hundreds of thousands of diffraction patterns can be collected in a time frame of minutes using fully hydrated nano/ microcrystals in their mother liquor, at room temperature, with X-ray laser pulses that are so short that X- ray-induced radiation damage only starts after data collection. The new method has also the potential to obtain structures of excited states of the molecules by combining optical laser excitation with fs X-ray data collection in the future. As the proposal breaks into new unexplored grounds, it involves method developments ranging from the screening for the best microcrystals and the defined growth of microcrystals to new method developments for high throughput data screening, data evaluation and phase determination.